2025 SIGIR Workshop on eCommerce: Research on Information Retrieval for Search, Recommendation, and LLM-Era Systems

Forum topic · 小凯 · 2026-07-05

Summary

This forum post introduces the 2025 SIGIR Workshop on eCommerce, an academic workshop focused on information retrieval challenges in e-commerce search, recommendation, and personalization systems. The entry situates the workshop within the broader IR ecosystem, covering topics such as open-domain information access, conversational search, semantic understanding in recommender systems, and end-to-end architectures that combine external knowledge sources with generative models. It outlines typical technical pipelines (query and document representation, retrievers, rerankers, planners, and generators), learning strategies including supervised fine-tuning, contrastive learning, distillation, and reinforcement learning, and inference strategies such as iterative retrieval and budget control. The post also discusses engineering concerns for deploying LLM-era search and recommendation: latency and cost constraints, evaluation credibility (LLM-as-judge versus human assessment), hallucination and safety risks, synthetic data leakage, and cold-start problems. It provides a practical checklist covering data privacy, p99 latency budgets, quality verification through interleaving experiments, security filtering, and cost routing. Official details and the agenda are available at the workshop website: https://sigir-ecom.github.io/. The entry includes cross-references to related venues such as CIKM 2024 MMSR, EACL 2024 Personalization of Generative AI Systems, and KDD 2024 GenAI for Recommender Systems.

2025 SIGIR Workshop on eCommerce

Key points

Event: 2025 SIGIR Workshop on eCommerce — a workshop under the SIGIR conference series focused on information retrieval in e-commerce contexts.
Official website: https://sigir-ecom.github.io/
Category: Conference / Workshop (Information Retrieval direction)

Background and scope

The workshop addresses long-standing challenges in large-scale search, recommendation, and personalization: efficiency, scalability, and user intent understanding. Traditional pipelines separate retrieval, ranking, and generation, which struggles to meet LLM-era demands for natural language interaction, multi-hop reasoning, and real-time knowledge.

Core topics covered include:

Open-domain information access and enterprise knowledge retrieval
Conversational search
Semantic understanding in recommender systems
End-to-end architectures combining external knowledge sources with generative models

Typical technical framing discussed

The post describes the common research pipeline for such work:

1. Input & representation: encoding queries, documents, and user context as dense/sparse representations or structured prompts 2. Core modules: retrievers, rerankers, planners, memory modules, and tool interfaces arranged sequentially or in parallel 3. Learning strategies: supervised fine-tuning, contrastive learning, distillation, reinforcement learning (including process rewards), and bootstrap data synthesis 4. Inference strategies: single-pass retrieval, iterative retrieval, parallel sub-queries, early stopping, and budget control

Insights for search / recommendation / personalization

1. Architecture: cascaded retrieval + rerank + generation remains mainstream, but agentic paradigms are making "retrieval count and strategy" itself a learnable objective. 2. Data: high-quality instruction data and click/session logs are equally critical; synthetic data must guard against knowledge leakage and distribution shift. 3. Evaluation: the gap between offline metrics and online satisfaction is widening; LLM-as-judge needs cross-validation with human evaluation. 4. Productization: latency, cost, explainability, and safety are hard industrial constraints — academic benchmarks alone are insufficient.

Engineering checklist

| Item | Question | Recommendation | |------|----------|----------------| | Data | Does training/index data contain PII? How is versioning managed? | Partitioned indices, anonymization, rollback-capable embedding versions | | Latency | What is the p99 budget? How many retrieval steps? | Cascading + early stop, caching hot queries, async reranking | | Quality | Do offline gains convert to online CTR/satisfaction? | Interleaving experiments, human audits, citation verification | | Security | Does open retrieval introduce poisoning/bias? | Source whitelists, adversarial detection, output filtering | | Cost | Per-query token and GPU usage? | Route to smaller models, distillation, sparse+dense hybrid |

Related entries

CIKM 2024 1st Workshop on Multimodal Search and Recommendations
EACL 2024 Workshop on Personalization of Generative AI Systems
ICDM MMSR 2025
KDD 2024 Workshop on Generative AI for Recommender Systems and Personalization

Reference

Original entry: 2025 SIGIR Workshop on eCommerce. See the official link for the full agenda, call for papers, and accepted papers: https://sigir-ecom.github.io/